Sketchup needs to be BIM
-
Think of BIM as a manual of a building, that can be referenced for the full life of that building. And in the early stages, a platform that can ensure that the building will perform up to the standards it is touted to.
We do that now, but it is all fragmented, and requires referencing outside the documentation of the building.
So if your a facilities manager now, and you have a mechanical problem. You go to the blueprints, find the reference to the mechanical page, and when you get to it you will find written words that tell you that the problem is a Honeywell A2446 pump. Trouble is the change order that authorized the Grundfoss pump your looking at isn't necessarily stapled to the page, so it's really worthless to you, so you begin your search.
In a BIM world the full set of electronic documents, would not only contain the page with the pump on it, but BIM management would have required the changeout documented, and when you drill down, the pump itself would be a full set of schematic diagrams, so you could get the part number for faulty O-ring and get it ordered.
Not necessarily a great example, but you get the picture.
But the statement that started this discussion is "SketchUp Needs to be BIM".My answer would be "Yes", if it plans to be a vital part of the Architectural Industry in the future. But I believe this to be a monster undertaking, and also may trigger a review of who the core users are, and if they, like Jeff, actually require BIM integration.
-
I am a "architectural engineer"(i hope it's correctly translated
, and in my spare time I try to make a BIM plugin for sketchup, but it would be great if the SketchUp team would make a real SketchUp BIM-addon(without losing sketchup's coolness and simplicity
)
I think BIM should be all about cooperation, between parties in a construction project AND between all pieces of software used to realize and manage a building.
Thinking in revit=BIM won't help anyone in the long run, because Autodesk is only interested in information exchange between different autodesk products.
I know there is a long way to go before information exchange using the IFC file format will be perfect, but I think IFC is the only way to go.SketchUp should not be cluttered with too many different IFC entities, just the basics. So a design made in SketchUp using these elements can be the base for a FULL building model hosted on some kind of BIM server for access by all project members(and softwares).
And even these basic elements can be very simple:
- Sketchup's "face" object is a perfect base for a "wall", a "floor", a "roof-slab". All you need is to add thickness and a few properties.
- Sketchup's edge object is a perfect base for all "columns", "beams" and window frames. Just add a profile and some properties.
It wouldn't even be necessary to add a separate button for all these elements. For me they are all "planars" and "linears".
For my plugin I try to stay as close to the sketchup-workflow as I can while adding the needed properties to export useful elements to IFC(and in the process to Revit, Archicad Tekla and whatever). But I still have a long way to go before it is really stable, the observers I use to monitor the base-geometry are still a bit tricky/unstable. I bet the SketchUp team could do a way better job!
If someone likes to check it out, especially the "workflow" (which I think is pretty sketchup-like), check out: http://sketchucation.com/forums/viewtopic.php?f=323&t=34007
-
@unknownuser said:
sawhack: It sounds like you're thinking of a set of new parametric entities (wall, floor, roof, window, door, ..?) with new tools to make them? Is there more to BIM than this?
John, I see a few additional things. Above you described the key planer BIM objects. We also need static objects, like casework, furniture, lights, etc. I think Dynamic components cover this well, but we hit some issues because they are built on ruby, and our plugin relies so heavily on observers. It would be nice to see that in C++ if that would make things more stable. The other important object is the "room" or "space" object. This is key for organizing the model.
In an OpenStudio model surfaces, lights and other objects are a child to a space, while in gbXML surfaces stand on their own but then refer to one or more spaces. I don't think either way is wrong or right, but just something that you will want to give a lot of thought. Then constructions, schedules, and light definitions (vs. an instance) are resource. Much like a component definition vs. instance. One limitation of working with OpenStudio models in SketchUp is that you have to work in a prescribed workflow. The more BIM-like SketchUp can be I think the more free the workflow can be.
I think another key BIM characteristic is Interoperability with other tools. For us that means gbXML, or direct translation between tools. Anyone can write translators to or from OpenStudio, and since it is OpenSource they can extend the model to add objects it is missing. For instance we support EnergyPlus and Radiance for energy and daylight analysis. If someone wants to write a translator to a model format for acoustic analysis, they will need to extend the model to for additional material properties. I think the attribute library in SketchUp generally supports this approach, it may just be formalizing the attributes more.
http://openstudio.nrel.gov/
http://www.youtube.com/user/NRELOpenStudioDavid
-
Hi there,
I'm a UK Architect and Master-planner, who has foregone the traditional CAD programs in recent years. I have designed extension to towns and extensions to listed buildings in SketchUp alone (Sir Walter Raleigh's mansion, none the less).
BIM as a 'process' is either helpful, or not depending on the result you are looking for. The St Martin's Centre, Cambridge, UK did a nice piece about the ^benefits^ of re-drawing a scheme as it passes key gateways. I favour this.
-
Sketch schemes and initial massing studies can be quite loose.
-
More detail studies,that might look at the 'spatial puzzle' that is a building, need to be more precise. I myself model at a scale, that on paper (yes, I did that too), would probably be 1:50 (maybe, 1:20). However, I do like the rigor of attributing data - i.e. taking off key metrics.
Personally, I end up 'simplifying' many BIM models, yet try to keep the attribute intact.
SketchUp is a wonderful workbench for designers of all flavours; if one 'corner' wins, by inference, the other looses! This would be a shame.
Garbage IN: Garbage OUT (GIGO) is the old saying. If BIM = less garbage = Great!
We have been working on a daylight and sunlight analysis plugin for SketchUp that is about 'less garbage' at the other end of the spectrum - the massing model end! This offer levers off SketchUp in the same way BIM does (by using sensible atributes to calculate lux levels, daylight factors, ADF, Vertical Sky Components and Insolation/Solar Gain).
Our hope is to turn SketchUp into 'the platform' for creating urban plans that facilitate low-energy construction - with a 40%-70% energy saving across larger scale development.
Again, SketchUP is an excellent spatial workbench. So, I only ask for 'less garbage', less 'tunnel vision' and point out that occasionally its is ^actually better^ to re-draw. In which case, everybody can have what they want!
-
-
[quote="brewsky"]I am a "architectural engineer"(i hope it's correctly translated
, and in my spare time I try to make a BIM plugin for sketchup, but it would be great if the SketchUp team would make a real SketchUp BIM-addon(without losing sketchup's coolness and simplicity
)
As a user of both Revit and Sketchup, I agree with Brewsky's idea of a BIM plugin that doesn't sacrafice Sketchup's coolness and simplicity. I think the idea of plug and play is a good one. To me trying to transform Sketchup into a Revit knock off, would be a huge mistake, because it would mean losing the things that make Sketchup great. Currently, Revit and Sketchup are two unique tools in my tool box that are used for different purposes. Sketchup is used for design and visulization, and Revit is used for documentaion, coordination, analysis, and basically taking a project from design development through CD's and final construction. To me Sketchup's current strengths lend themselves to the beginning phases of a project, and Revit to the later phases. Given the choice between the two I would choose Sketchup hands down when it comes to modeling and visualization. It is more inuititive, faster, and easier to learn and use. If I am on a Revit project and have to produce renderings, or model design options, I often export my Revit model and bring it into Sketchup. I can model 3 design options for a project, to a nice rendered quality, much quicker in Sketchup than I can in Revit. But it is hard to make those modeling efforts pay off after one of the design options is chosen and we move to construction documents. As others have said we need to find ways to make the Sketchup model more beneficial down the pipeline in the design process.
There is so much discussion and much confusion on what BIM is. I have had wade through a lot of it in order to help our partners, colleagues, and clients understand. There is BIM the tool and there is BIM the process. The process of BIM, which has been described as the 'Integrated Design Process', is as much more about a different way of thinking and doing architectural design than it is about software. While the goal of a more seamless process, and more integrated building are the goal, both the tool and the process have been very hard to implement. Revit as a tool is not intuitive, there are many menus and schedules to drill through. It requires heavy front end loading. It is very data driven. It is challenging for visual thinking, intuitive designers to adopt to this kind of tool. I have modeled projects, using with SketchUp alone, that have been very integrated in terms of the design process and how Sketchup was used on the project. We used the model for design purposes, we did take offs, we did analysis, we coordinated building systems to some degree, we even exported views of the model into AutoCAD line work for CD's, and we used the model to help the contractor understand and building during construction. BIM does not equal Revit. My hope is the Sketchup / Trimble team come up with something very different than Revit, that builds on Sketchup's strengths, and that approaches the process of BIM much more intuitive, way. A good first step would be to make the going from Sketchup to Revit and vice versa much more seemless. That would allow Sketchup models developed in early design to have more of a payoff through out the process.
-
@jbacus said:
This is an interesting topic for me, and it speaks to a number of other discussions I have had around the internet lately. If SketchUp were to be "BIM", what entities, tools or other features would have want it to have?
sawhack: It sounds like you're thinking of a set of new parametric entities (wall, floor, roof, window, door, ..?) with new tools to make them? Is there more to BIM than this?
john
.Glad to see this an ongoing topic even if under new threads... well done sawhack for starting it off!
@ John B - I think parametric tools would be a major leap forward into the architectural field. What Aaron and his team at BuildEdge are doing with their plugin is very impressive but for SketchUp Pro for Architect [for want of a title] I see a set of tools as such:
Wall tool
Dialogue box info:-
inputs for wall thickness / height
-
Wall position in relation to input. [left, right, centre or custom offset]
-
Wall junction options - mitre, butt joint [this allows for individual wall segments to be edited if needed]
When drawn, each wall segment would be a group therefore allowing individual editing of wall height to u/s of roof, etc or raking the wall top to meet roof planes...
Wall Edits
- a offshoot of solid tools whereby the user can trim this to that or extend this to that.... as Aidan explains it!
Opening tool - Doors / windows
Dialogue box info:-
Window type [I envisage each user creating his / her own custom component which forms the window configuration. Data below is used to scale it to the desired size]
-
inputs for width / height
-
Wall position in relation to input. [left top corner, right bottom, etc]
-
Wall thickness the opening is being placed in [required to cut a hole in the wall [solid] as does the W x H data...]
Opening Edits
- would be nice to be able to click opening edits then select the opening and the dialogue box appear from which the user had input the data. From here we can change what we want even change the type of window configuration and save out...
Same for columns, roofs, stairs, yes landscape tools also! Cut N fill tools as an extension to the stamp tool for example....
Layers
Layers are the other thing that need to be re-worked in order for SketchUp to be more BIM like. Combining Outliner and the Layer manager to be able to nest layers would be a great start!As for a 2D UI, I'm not so concerned. SketchUp is what it is because of its unique workspace. You can still work in 3D aerial and model if you want 2D....
What I do think should be enhanced along with these ideas is the ability for Layout to find boundary searches of geometry for hatching, filling, etc. Many more things to add for Layout ideas but need to find the time to write them!!
I'll be back to this thread tonight or over the weekend hopefully to add more.... such an exciting topic!!!!!
-
-
The last few posts are extremely interesting, with great ideas.
But I can't help but pose the question... Do you think it is feasible to be only partially BIM integrated? A kind of BIM lite? -
@dale said:
Do you think it is feasible to be only partially BIM integrated? A kind of BIM lite?
Every bit of useful / meaningfull data that can be attributed to a building model (in any part of the building process) makes your BIM better, don't you think?
Making an energy calculation won't use all your BIM-data either, it's just part of the process...I wouldn't speak of BIM-lite, but I think SketchUp can be a very useful part in a larger OPEN-BIM process.
Just create the basic building model using sketchup(new building parts), upload it to a bim-server and expand on it using all other kinds of software(construction(tekla?), MEP software).
I really like the approach of http://bimserver.org/ I'm trying to figure how to make a connection with that at the moment...
-
I've used ACAD for years. I was trained on version 2.3 and used it until 2010 (yikes that's 24 years if you do the math), when I started my switch to all SU. Depending on how you organize your modeling in SU, there are many aspects of SU that inherently create a BIM model. The components and how they are tracked are wonderful in maintaining instance count as well as descriptive information. The fact that SU works off planar geometry is also a powerful tool for quantity take offs, by area or material. If SU could then quantify this information into simple schedules, that would really make a giant step for the BIM community.
That said, the last thing I want to see is SU turn anything close to the beast that is Revit. I review lots of other Architect's work here as I serve on a design review board. I can't tell you how many times I see inaccuracies in the use of their BIM software. The one complaint I here is the learning curve and lack of intuitive control is not what they had expected. These two elements alone are probably the greatest attributes of SU.
I want to see something different and unique. The technology is there and it is amazing what is being developed by several companies based around SU. BIM has such a broad meaning and it certainly shouldn't be confined to what Revit produces.
-
@unknownuser said:
Depending on how you organize your modeling in SU, there are many aspects of SU that inherently create a BIM model. The components and how they are tracked are wonderful in maintaining instance count as well as descriptive information. The fact that SU works off planar geometry is also a powerful tool for quantity take offs, by area or material.
I like what you say here.
-
definitely not off topic but I'll label it as such anyway..
[off:21a0jjy1]
@brewsky said:@unknownuser said:
The fact that SU works off planar geometry is also a powerful tool for quantity take offs, by area or material.
I like what you say here.
<<well, aside from the fact that a 12' rad sphere in sketchup reports about 8 cu/yd less concrete than what's really needed.. (or around 25 sqft less surface area) >>
just a friendly reminder to be careful when using volume and area reports in sketchup on non-box shapes[/off:21a0jjy1]
-
@unknownuser said:
definitely not off topic but I'll label it as such anyway..
[off:2bp7mwqv]
@brewsky said:@unknownuser said:
The fact that SU works off planar geometry is also a powerful tool for quantity take offs, by area or material.
I like what you say here.
<<well, aside from the fact that a 12' rad sphere in sketchup reports about 8 cu/yd less concrete than what's really needed.. (or around 25 sqft less surface area) >>
just a friendly reminder to be careful when using volume and area reports in sketchup on non-box shapes[/off:2bp7mwqv]
I hear you Jeff. Thankfully I'm not designing any concrete spheres! I guess if I were, they would be more cost issues in the forms than the $1400 difference in concrete.
-
@dale said:
The last few posts are extremely interesting, with great ideas.
But I can't help but pose the question... Do you think it is feasible to be only partially BIM integrated? A kind of BIM lite?Baby steps at first, Dale.....
-
BIM is an undefined acronym. Its boundaries are so ill defined I'm surprised people are still trying to put it in a box. It is an impossible concept for all but the most meticulous and open funded projects. The reality is we (of an architectural persuasion) are just seeking tools to make the construction process easier. The prospect of being able to create data and pass it around to other software users and expect them to be able to interrogate the data in a meaningful way other than for basic shapes is not going to happen due to basic commercial posturing. If Trimble want to take SU down the BIM route it should recognise these challanges and offer, (as some have mentioned above) the ability to make realistic interrogations that can be output in a useful way. That shouldn't be a major challenge as the basic scheduling tool is there in the "Report Generator". What needs to be improved is the accessibility and accuracy of that data to give 2D/3D Perimeter / Area / volumes and improvements to the list presentation. I am far from convinced that there is anything to be gained from adding data tags beyond the already present component decription as in the majority of cases it will be a complete waste of time for real world construction projects.
IMHO.
-
Stay tuned I have an answer to all of your questions and more.
-
You're a tease!!!!!!
-
I work with a local contractor that has a robotic plasma cutter. We designed several specific wood/steel connections solely in SU. I then exported .dxf files from SU for input in their plasma cutter software. It was remarkable how seamless the process was, and the results were nearly immediate - 3 days. I'll try to find some pics of the process, but it clearly represents the BIM capabilities already available in SU.
-
The BIM thing has been intriguing me lately. One of the other (!) software companies has been pushing their software to create BIM models, but having watched their videos, I felt that SU could do most of what was required,i.e. 1) a 3d model 2) able to render from the model 3) using 3d components 4) extracting component information for estimating. It was only the last item that I found a bit flaky. (I know that there are timelines and facilities management stages, but that isn't necessary for my domestic work)
For small, domestic projects a full-on BIM (using sophisticated layering conventions etc) seems way over the top, but it did occur to me that to assist the builder in pricing for the work, that I could lend a hand and extract areas and volumes of materials, lengths of beams etc, that a builder, without the benefit of the 3d model would otherwise have to prepare manually. The only danger is, me making an error!, so it does put the onus on me to double check the output. (most small builders don't have CAD software anyway, let alone a computer)
I've just given this a try out on my last project, preparing "QUANTITIES of Main Elements
(from model)" and with a hefty health warning about the data being to assist the builder etc. I will wait for feedback from the builder quoting for the work. -
That sounds neat, Nick... would love to see the process pics...!
-
@utiler said:
That sounds neat, Nick... would love to see the process pics...!
The piece is a center steel knife plate with 8 plates. It is 54" in diameter. In the first video, that google released on my process, there is a small segment showing the model of this element.
No shop drawings were prepared other than the SU model.
Here it is freshly installed.
Advertisement